CN109521029A - A kind of detection method of bayonet type automobile lamp lamp cap side open defect - Google Patents
A kind of detection method of bayonet type automobile lamp lamp cap side open defect Download PDFInfo
- Publication number
- CN109521029A CN109521029A CN201811453963.1A CN201811453963A CN109521029A CN 109521029 A CN109521029 A CN 109521029A CN 201811453963 A CN201811453963 A CN 201811453963A CN 109521029 A CN109521029 A CN 109521029A
- Authority
- CN
- China
- Prior art keywords
- lamp cap
- cap side
- line
- lamp
- edge
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/94—Investigating contamination, e.g. dust
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
Abstract
The invention proposes a kind of detection methods of bayonet type automobile lamp lamp cap side open defect.Step of the present invention: one, bayonet type automobile lamp lamp cap side image obtains；Two, lamp cap lateral side regions position: including using canny operator detection lamp cap lateral side regions edge and hough transformation to find line segment；Three, lamp cap side open defect is detected: dirty including detection lamp cap side and weigh wounded, torticollis defect, marking defects and Bian Si open discontinuity.The present invention have many advantages, such as high degree of automation, without manual intervention, detection speed it is fast, detection accuracy is high, can detect number of drawbacks simultaneously.
Description
Technical field
The present invention relates to the inspections of field of visual inspection more particularly to a kind of bayonet type automobile lamp lamp cap side open defect
Survey method.
Background technique
Automobile lamp is the essential element of automobile, primarily serves illumination and the effect of signal designation, driver is allow to exist
Night and overcast and rainy greasy weather safe driving.Bayonet base bulb is widely used in car light, and the light bulb of the type generally includes three parts:
Glass shell (abbreviation glass bulb), the intracorporal filament of lamp and the lamp cap of lamp body.The surface of lamp cap side is printed on word, identifies light bulb
Relevant information, convenient for being identified when car light assembly.It is mechanical due to the structure and its produced in series process of bayonet type automobile lamp complexity
The precision of control system limits, and various defects inevitably occurs in light bulb product.Bayonet type automobile lamp lamp cap side is main
Detection project be lamp cap side it is dirty weigh wounded, whether there is or not exposed four classes of, side silk for torticollis, lettering.These possible defects are subsequent
It will lead to serious safety problem in car light assembling and driving use.Therefore, the necessary compulsory test light bulb before car light factory
Open defect.
Currently, for bayonet type headlight appearance defect detection there is still a need for artificial range estimation processing, that is, be arranged multiple and different
Detection station introduces a large amount of lamp inspection works and carries out artificial detection.However, artificial detection has been unable to meet Product quality and safety and high speed
The requirement of mass automatic production, is mainly manifested in: (1) artificial detection precision is low, and subjectivity is strong, and repeatability is poor, can not
It provides and continues consistency detection as a result, missing inspection false detection rate is high；(2) increasing for detection project steps up with examination criteria, manually
In the presence of detection blind spot；(3) manual detection efficiency is low, great work intensity, and a worker can only detect 500-800 per hour, hardly possible
To meet production line high-speed production requirement.Therefore, the headlight appearance defect test technique automatic tool of research substitution artificial detection
There is great economy and society meaning.
The present invention is based on current machine vision-based detection principle, for bayonet type automobile lamp lamp cap side difference appearance
Defect proposes different visible detection methods, to meet the diversified demand of headlight quality detection.The detection that the present invention researches and develops
Current manual's examination criteria is fully achieved in method in terms of detection accuracy, accuracy and reliability, and has in terms of detection efficiency
There is huge advantage.
Summary of the invention
In view of the deficiencies of the prior art, the present invention proposes a kind of detection of bayonet type automobile lamp lamp cap side open defect
Method, this method can efficiently detect bayonet type automobile lamp lamp cap side open defect.
The present invention is realized by the following scheme:
Step 1: bayonet type automobile lamp lamp cap side image obtains；Lamp cap side includes 360 ° of models for surrounding lamp cap
It encloses, since the areas imaging of a camera is limited, therefore uses four and be looped around the camera of lamp cap surrounding to lamp cap image side faces；For
The strip source of one group of each camera design at an angle devises low angle as side light source above lamp cap
Annular light source, acquired image is known as dark figure I when side light source is closed and top annular light source is openeddark, side light source and
The image that top annular light source obtains when opening simultaneously is known as bright figure Ibright；
Step 2: to bright figure IbrightIn lamp cap lateral side regions positioning；Including using canny operator extraction lamp cap side
Line segment is found in edges of regions and use hough transformation；
Step 3: being detected to lamp cap side open defect；It is right including dirty to lamp cap side and weigh wounded and detect
The torticollis defect of lamp cap is detected, and is detected to lamp cap side marking defects, is carried out to lamp cap side face edge silk open discontinuity
Detection.
The invention has the following advantages that
1, the present invention utilizes machine vision technique, carries out to the open defect of bayonet type automobile lamp lamp cap side full-automatic
Detection is not necessarily to manual intervention, largely saves human resources；
2, the present invention can simultaneously detect number of drawbacks, and detection speed is fast, and precision is high, and system is reliable and stable.
Detailed description of the invention
Fig. 1 is implementation flow chart of the invention；
Fig. 2 is bayonet type automobile lamp lamp cap detection zone schematic diagram；
Fig. 3 is the testing result display diagram in one group of bayonet type automobile lamp lamp cap left side face to be detected；
Fig. 4 is lamp cap side region of search setting schematic diagram；
Fig. 5 is the positioning schematic diagram of lamp cap side；
Fig. 6 is dirty and weighs detection schematic diagram wounded；
Fig. 7 is to expand signal schematic representation；
Fig. 8 is peak detection result schematic diagram；
Fig. 9 is the angle schematic diagram at glass bulb edge Yu lamp cap edge；
Specific embodiment
Embodiment 1
Bayonet type automobile lamp lamp cap side includes 360 ° of ranges for surrounding lamp cap, due to the areas imaging of a camera
It is limited, therefore use four cameras for being looped around lamp cap surrounding to lamp cap image side faces, the image point that four cameras are obtained respectively
Analysis judges lamp cap side with the presence or absence of open defect based on the analysis results.
It is illustrated by taking the face to be detected of first side of bayonet type automobile lamp lamp as an example, the detection process in other faces to be detected
It is consistent therewith.Bayonet type automobile lamp lamp cap detection zone is as shown in Figure 2.
As shown in figure 3, illustrating the testing result in one group of first face to be detected of bayonet type automobile lamp lamp.
In conjunction with Fig. 1, of the invention the specific implementation steps are as follows:
Step 1: bayonet type automobile lamp lamp cap side image obtains:
It highlights lamp cap edge (including side silk detection, lamp cap edge, glass bulb edge) due to be able to satisfy and prevents lamp body
The reflective requirement of glass surface can choose annular light source or strip source at an angle, consider further that lamp cap side is dirty
Dirty little with the feature for weighing direct imaging wounded and background difference, the high brightness for forming interval in base surface using striped light source is bright
Striped and dark fringe, appearance that is dirty or weighing wounded can change light and shade striped, i.e. the gray scale of its corresponding position mutates, so as to
Will be dirty or weigh wounded and distinguished with background, therefore the present invention designs following imaging schemes to obtain lamp cap side image:
Lamp cap side includes 360 ° of ranges for surrounding lamp cap, since the areas imaging of a camera is limited, therefore uses four
A camera for being looped around lamp cap surrounding is to lamp cap image side faces；For the strip source of one group of each camera design at an angle
As side light source, and low angle annular light source is devised above lamp cap, side light source is closed and top annular light source is beaten
Acquired image is known as dark figure I when openingdark, the image that side light source and top annular light source obtain when opening simultaneously is referred to as bright
Scheme Ibright；
In the present embodiment, the dark figure I in the bayonet type automobile lamp lamp cap left side face to be detected that camera obtainsdarkWith bright figure
Ibright(Fig. 3 b, Fig. 3 d, Fig. 3 e are dark figure I as shown in Figure 3dark, other are bright figure Ibright)；
Step 2: to bright figure IbrightIn lamp cap lateral side regions positioning:
1) in bright figure IbrightIn, it is calculated in region of search with canny for the bottom edge setting region of search of lamp cap side
Son extracts lamp cap side bottom edge edge, then finds line segment using hough transformation, the endpoint for obtaining the line segment obtains the line segment
Endpoint (x1,y1) and (x2,y2), straight line Line where the line segment1Equation be expressed as under image coordinate system:
Using the straight line as the bottom edge of lamp cap side；
2) region of search is set for the top margin of lamp cap side, in region of search, with canny operator extraction lamp cap side
Then top margin edge finds line segment using hough transformation, obtains the endpoint (x of the line segment3,y3) and (x4,y4), line segment place
Straight line Line2Equation be expressed as under image coordinate system:
Using the straight line as the top margin of lamp cap side；
3) region of search is set for the right of lamp cap side, in region of search, with canny operator extraction lamp cap side
Then the right edge finds line segment using hough transformation, obtains the endpoint (x of the line segment5,y5) and (x6,y6), line segment place
Straight line Line3Equation be expressed as under image coordinate system:
Using the straight line as the right of lamp cap side；
4) above-mentioned three straight lines intersect two-by-two, simultaneous straight line Line1And Line3Solution of equation obtain straight line Line1And Line3
Intersection point A coordinate be (xA,yA), simultaneous straight line Line2And Line3Solution of equation obtain Line2And Line3Intersection points B coordinate
For (xB,yB)；
5) coordinate (x of the another summit D on lamp cap side bottom edgeD,yD) are as follows:
Coordinate (the x of the another summit C of lamp cap side top marginC,yC) are as follows:
In the present embodiment, lamp cap side region of search setting such as Fig. 4, the positioning of lamp cap side such as Fig. 5；
Step 3: being detected to lamp cap side open defect:
1) dirty to lamp cap side and weigh wounded and detect:
1. as shown in fig. 6, in bright figure IbrightIn, for each striped, preset several horizontal line ai(i=1,
2 ..., n), straight line aiOn pixel and vertical direction ωa(3≤ωa≤ 7) pixel is cumulative in neighborhood, is extended to new one-dimensional letter
Number Fi, the starting point of signal is 1, signal length n2-n1；
2. detecting signal F with multiple dimensioned gaussian shape probe methodiIn peak value:
Using σ as scale parameter, guarantee scale invariability using normalization coefficient γ=1；
In peak detection:
hr(x, σ)=- h (x, σ)
During valley detection:
Scale σminAnd σmaxRespectively correspond the half of defect width minimum and maximum in image, FiAnd its peak detection result
Respectively as shown in Fig. 7 and Fig. 8；
3. to signal FiIn peak value according to threshold value, determine car light with the presence or absence of dirty and weigh wounded；
In the present embodiment, dirty testing result as illustrated in figure 3f, weighs testing result wounded as shown in figure 3g；
2) lamp cap side torticollis defect is detected:
1. in bright figure IbrightIn, it is calculated in region of search with canny for car light glass bulb top edge setting region of search
Son extracts lamp cap lateral edge, then finds line segment using hough transformation, obtains the endpoint (x of the line segment7,y7) and (x8,y8)；
2. as shown in figure 9, determining the angle γ at glass bulb edge and lamp cap edge:
γ=180 °-γ1-γ2
γ1=arctan | y8-y7/x8-x7|
γ2=arctan | y4-y3/x4-x3|；
3. the angle γ to glass bulb edge and lamp cap edge uses threshold decision method, determine that car light is lacked with the presence or absence of torticollis
It falls into；
In the present embodiment, torticollis defects detection result is as shown in Figure 3a；
3) lamp cap side marking defects are detected:
1. in secretly figure IdarkIn, detection zone is set for car light lamp cap lettering region, with median filtering to detection zone
It is filtered to obtain area to be tested；
2. counting intensity profile situation, gray level L to area to be testedi=m (i=0,1,2 .., 255 and 0≤m≤
255), m is a certain gray value, and corresponding pixel quantity is
3. the gray level of lettering part is much higher than the gray level of background parts, and concentrates at a certain specific gray value, right
Li> Tword (TwordFor standard lettering gray level) the corresponding pixel quantity of gray level be maximized；
4. being to pixel quantityUsing threshold decision method, determine that car light whether there is marking defects, i.e., (ST.Li> Tword) when zero defect；
Threshold value TwordAnd TnumIt is determined by experiment in advance: taking the standard lamp head of 100 intact letterings, each lamp cap imaging 5
It is secondary, it obtains 500 width and secretly schemes；By above-mentioned process, median filtering and gray scale are carried out to the detection zone of the lamp cap lettering of each image
Value side's figure statistics obtains gray level in jth width image larger (corresponding lettering) and the most gray level L of its respective pixel quantityj
(j=1,2 ..., 500) and its pixel quantity NumLj, one is obtained 500 groups of data；To the L in 500 groups of datajAnd NumLjPoint
It is not averaged to obtainWithIn view of lamp cap lettering itself and its imaging difference, need to there are certain surpluses, finally
By threshold value TwordAnd TnumIt is defined respectively as:
In the present embodiment, marking defects testing result is as shown in Figure 3b；
4) lamp cap side face edge silk open discontinuity is detected:
1. in bright figure IbrightIn, for car light side silk setting region of search, in region of search, with canny operator extraction
Then line segment is found using hough transformation in lamp cap and glass bulb junction edge, determine in region of search with the presence or absence of side silk；
2., using SURF algorithm to lamp cap and glass bulb junction extracted region spot, obtaining spot face if there is side silk
Product；
A. Hessian matrix is constructed:
Wherein, LxxThe integral of (x, σ) expression Gauss second dervativeL can similarly be obtainedxy(x, σ) and Lyy(x,σ)；
The approximation of the Hessian matrix determinant of each pixel:
det(Happrox)=Dxx*Dyy-(0.9*Dxy)2
B. scale pyramid iconic model is constructed, the pyramid model in SURF algorithm is big by changing Gaussian filter
It is small, it carries out Gaussian Blur and obtains every layer of picture to be detected；
C. characteristic point is primarily determined using non-maxima suppression:
By 3 dimension neighborhoods in each pixel and its scale pyramid iconic model crossed by hessian matrix disposal
26 points carry out size comparison remain, if it is the maximum value or minimum value in this 26 points as preliminary
Characteristic point；
D. it is accurately positioned extreme point:
The characteristic point of sub-pixel is obtained using 3 dimensional linear interpolation methods, while also removing those values less than certain threshold value
Point；
E. the harr wavelet character in statistical nature vertex neighborhood, the principal direction of selected characteristic point；
F. feature point description is generated:
A square-shaped frame is taken around characteristic point, the side length of frame is that (s is the ruler where detected this feature point to 20s
Degree), the frame is then divided into 16 sub-regions, each subregion counts the haar horizontally and vertically of 25 pixels
Wavelet character obtains feature vector, i.e. feature point description；
G. the Euclidean distance and Hessian trace of a matrix between two characteristic points are calculated, the matching of characteristic point, institute's Matching band are reached
The domain region Ji Biansi extracts side silk spot, obtains speck area；
3. using threshold decision method to speck area, determine car light with the presence or absence of side silk open discontinuity；
In the present embodiment, side silk open discontinuity testing result is as shown in Figure 3c.
Claims (3)
1. a kind of detection method of bayonet type automobile lamp lamp cap side open defect, the method includes at least following step
It is rapid:
Step 1: obtaining bayonet type automobile lamp lamp cap side image, including the dark image of a width and a bright image；
Step 2: in image lamp cap lateral side regions position: including use canny operator detection lamp cap lateral side regions edge and
The line segment found on lamp cap side bottom edge, top margin and right side is converted using hough；
Step 3: being detected to lamp cap side open defect: including detection lamp cap side it is dirty and weigh wounded, torticollis defect, print
Word defect and Bian Si open discontinuity.
2. the detection method of bayonet type automobile lamp lamp cap side according to claim 1 open defect, which is characterized in that
The step 2 specifically includes the following steps:
1) region of search is set for the bottom edge of lamp cap side, in region of search, with the operator extraction lamp cap side canny bottom edge
Then edge finds line segment in the edge of extraction using hough transformation, obtains the endpoint (x of the line segment1,y1) and (x2,y2),
Straight line Line where the line segment1Equation be expressed as under image coordinate system:
Using the straight line as the bottom edge of lamp cap side；
2) region of search is set for the top margin of lamp cap side, in region of search, with canny operator extraction lamp cap side top margin
Then edge finds line segment using hough transformation, obtains the endpoint (x of the line segment3,y3) and (x4,y4), straight line where the line segment
Line2Equation be expressed as under image coordinate system:
Using the straight line as the top margin of lamp cap side；
3) region of search is set for the right of lamp cap side, in region of search, on the right of canny operator extraction lamp cap side
Then edge finds line segment using hough transformation, obtains the endpoint (x of the line segment5,y5) and (x6,y6), straight line where the line segment
Line3Equation be expressed as under image coordinate system:
Using the straight line as the right of lamp cap side；
4) above-mentioned three straight lines intersect two-by-two, simultaneous straight line Line1And Line3Solution of equation obtain straight line Line1And Line3Intersection point
The coordinate of A is (xA,yA), simultaneous straight line Line2And Line3Solution of equation obtain Line2And Line3Intersection points B coordinate be (xB,
yB)；
5) coordinate (x of the another summit D on lamp cap side bottom edgeD,yD) are as follows:
Coordinate (the x of the another summit C of lamp cap side top marginC,yC) are as follows:
So that it is determined that the detection zone of lamp cap side.
3. the detection method of bayonet type automobile lamp lamp cap side according to claim 1 open defect, which is characterized in that
The step 3 specifically includes the following steps:
1) dirty to lamp cap side and weigh wounded and detect:
1. in the picture, presetting several horizontal line a for each stripedi(i=1,2 ..., n), straight line aiOn picture
Element and vertical direction ωa(3≤ωa≤ 7) pixel is cumulative in neighborhood, is extended to new one-dimensional signal Fi, the starting point of signal is 1, letter
Number length is n2-n1；
2. detecting signal F with multiple dimensioned gaussian shape probe methodiIn peak value；
3. to signal FiIn peak value use threshold decision method, determine car light with the presence or absence of dirty and weigh wounded；
2) lamp cap side torticollis defect is detected:
1. region of search in the picture, is arranged for car light glass bulb top edge, in region of search, with canny operator extraction vehicle
Then lamp glass bulb top edge finds line segment using hough transformation, obtains the endpoint (x of the line segment7,y7) and (x8,y8)；
2. determining the angle γ at glass bulb edge and lamp cap edge:
γ=180 °-γ1-γ2
Wherein, γ1=arctan | y8-y7/x8-x7|
γ2=arctan | y4-y3/x4-x3|；
3. the angle γ to glass bulb edge and lamp cap edge uses threshold decision method, determine that car light whether there is torticollis defect；
3) lamp cap side marking defects are detected:
1. detection zone in the picture, is arranged for car light lamp cap lettering region, detection zone is filtered with median filtering
Obtain area to be tested；
2. counting intensity profile situation, gray level L to area to be testedi=m (i=0,1,2 .., 255 and 0≤m≤255), m is
A certain gray value, corresponding pixel quantity are
3. the gray level of lettering part is much higher than the gray level of background parts, and concentrates at a certain specific gray value, to Li>
Tword(TwordFor standard lettering gray level) the corresponding pixel quantity of gray level be maximized；
4. being to pixel quantityUsing threshold decision method, determine that car light whether there is marking defects；
4) lamp cap side face edge silk open discontinuity is detected:
1. in the picture, for car light side silk setting region of search, in region of search, with canny operator extraction lamp cap and glass
Then line segment is found using hough transformation in shell junction edge, determine in region of search with the presence or absence of side silk；
2., using SURF algorithm to lamp cap and glass bulb junction extracted region spot, obtaining speck area if there is side silk；
3. using threshold decision method to speck area, determine car light with the presence or absence of side silk open discontinuity.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811453963.1A CN109521029B (en) | 2018-11-30 | 2018-11-30 | Method for detecting appearance defects of side surface of lamp holder of bayonet type automobile lamp |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811453963.1A CN109521029B (en) | 2018-11-30 | 2018-11-30 | Method for detecting appearance defects of side surface of lamp holder of bayonet type automobile lamp |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109521029A true CN109521029A (en) | 2019-03-26 |
| CN109521029B CN109521029B (en) | 2021-07-13 |
Family
ID=65794103
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811453963.1A Active CN109521029B (en) | 2018-11-30 | 2018-11-30 | Method for detecting appearance defects of side surface of lamp holder of bayonet type automobile lamp |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109521029B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021183853A1 (en) * | 2020-03-13 | 2021-09-16 | Applied Materials, Inc. | Apparatus and method for inspecting lamps |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20110103088A (en) * | 2010-03-12 | 2011-09-20 | 신용호 | Smoke test having perceivealbe that sampling probe was removed from exhaust pipe |
| CN102221557A (en) * | 2011-03-31 | 2011-10-19 | 南京航空航天大学 | Automatic detecting device and detection method for quality of lamp holder based on machine vision |
| CN203929012U (en) * | 2014-01-21 | 2014-11-05 | 图灵视控（北京）科技有限公司 | Automobile lamp lamp pearl detection system |
| CN104698010A (en) * | 2013-12-09 | 2015-06-10 | 三星电机株式会社 | Device used to detect appearance of chip assembly |
| JP2016142525A (en) * | 2015-01-29 | 2016-08-08 | 京セラ株式会社 | Light source for painted surface inspection, and luminaire for painted surface inspection |
| CN207516273U (en) * | 2017-11-20 | 2018-06-19 | 泰兴市新永诚车灯塑件有限公司 | A kind of automobile lampshade detection device |
| CN108564631A (en) * | 2018-04-03 | 2018-09-21 | 上海理工大学 | Car light light guide acetes chinensis method, apparatus and computer readable storage medium |
-
2018
- 2018-11-30 CN CN201811453963.1A patent/CN109521029B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20110103088A (en) * | 2010-03-12 | 2011-09-20 | 신용호 | Smoke test having perceivealbe that sampling probe was removed from exhaust pipe |
| CN102221557A (en) * | 2011-03-31 | 2011-10-19 | 南京航空航天大学 | Automatic detecting device and detection method for quality of lamp holder based on machine vision |
| CN104698010A (en) * | 2013-12-09 | 2015-06-10 | 三星电机株式会社 | Device used to detect appearance of chip assembly |
| CN203929012U (en) * | 2014-01-21 | 2014-11-05 | 图灵视控（北京）科技有限公司 | Automobile lamp lamp pearl detection system |
| JP2016142525A (en) * | 2015-01-29 | 2016-08-08 | 京セラ株式会社 | Light source for painted surface inspection, and luminaire for painted surface inspection |
| CN207516273U (en) * | 2017-11-20 | 2018-06-19 | 泰兴市新永诚车灯塑件有限公司 | A kind of automobile lampshade detection device |
| CN108564631A (en) * | 2018-04-03 | 2018-09-21 | 上海理工大学 | Car light light guide acetes chinensis method, apparatus and computer readable storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 王佳炎: "基于机器视觉的车用灯泡检测系统研究", 《中国优秀硕士学位论文全文数据库 工程科技II辑》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021183853A1 (en) * | 2020-03-13 | 2021-09-16 | Applied Materials, Inc. | Apparatus and method for inspecting lamps |
| US11460413B2 (en) | 2020-03-13 | 2022-10-04 | Applied Materials, Inc. | Apparatus and method for inspecting lamps |
| US11761901B2 (en) | 2020-03-13 | 2023-09-19 | Applied Materials, Inc. | Apparatus and method for inspecting lamps |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109521029B (en) | 2021-07-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112508826B (en) | Printed matter defect detection method | |
| CN106296666B (en) | A kind of color image removes shadow method and application | |
| CN111239158A (en) | Automobile instrument panel detection system and detection method based on machine vision | |
| CN108416766B (en) | Double-side light-entering type light guide plate defect visual detection method | |
| CN109724990A (en) | The quick positioning and detection method in coding region in a kind of packing box label | |
| CN109993154B (en) | Intelligent identification method for single-pointer sulfur hexafluoride instrument of transformer substation | |
| CN108596880A (en) | Weld defect feature extraction based on image procossing and welding quality analysis method | |
| CN109993099A (en) | A kind of lane line drawing recognition methods based on machine vision | |
| CN112577969B (en) | Defect detection method and defect detection system based on machine vision | |
| CN110108712A (en) | Multifunctional visual sense defect detecting system | |
| CN109886960A (en) | The method of glass edge defects detection based on machine vision | |
| CN111539927B (en) | Detection method of automobile plastic assembly fastening buckle missing detection device | |
| CN111667475B (en) | Machine vision-based Chinese date grading detection method | |
| CN104597057A (en) | Columnar diode surface defect detection device based on machine vision | |
| CN108844961A (en) | A kind of temperature controller case vision detection system and method | |
| CN110189375A (en) | A kind of images steganalysis method based on monocular vision measurement | |
| CN111060442A (en) | Oil particle detection method based on image processing | |
| CN112819844A (en) | Image edge detection method and device | |
| CN112304957A (en) | Machine vision-based intelligent detection method and system for appearance defects | |
| CN114119591A (en) | Display screen picture quality detection method | |
| CN114387269B (en) | Fiber yarn defect detection method based on laser | |
| CN108414531A (en) | A kind of fexible film defect detecting device and its detection method based on machine vision | |
| CN111426693A (en) | Quality defect detection system and detection method thereof | |
| Lin et al. | Surface defect detection of machined parts based on machining texture direction | |
| CN113139943B (en) | Method and system for detecting appearance defects of open circular ring workpiece and computer storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |